Re:[SIESTA-L] openmp

2019-02-07 Por tôpico 郑仁慧
Dear Nick : Thank you for your help. Ren-hui Zheng 发件人:Nick Papior 发送日期:2019-02-06 05:15:20 收件人:siesta-l 主题:Re: [SIESTA-L] openmp In the Siesta 4.1 manual there is a clear description (Sec. 2.3.2) on how to utilize hybrid MPI+OpenMP using OpenMPI. For your architecture

Re: [SIESTA-L] openmp

2019-02-06 Por tôpico Nick Papior
-04 15:07:10 > To: siesta-l > Subject: Re: [SIESTA-L] openmp > > Dear Ren-hui Zheng, > > Siesta can benefit from MPI, OpenMP or both. > The benefit may be briefly summarized as follows: > MPI has distribution across orbitals, hence a basic advice is that you get > perfo

Re:[SIESTA-L] openmp

2019-02-05 Por tôpico 郑仁慧
: [SIESTA-L] openmp Dear Ren-hui Zheng, Siesta can benefit from MPI, OpenMP or both. The benefit may be briefly summarized as follows: MPI has distribution across orbitals, hence a basic advice is that you get performance scaling with # of cores == # of atoms (rule of thumb) OpenMP also distributes

Re: [SIESTA-L] openmp

2019-02-04 Por tôpico Nick Papior
Dear Ren-hui Zheng, Siesta can benefit from MPI, OpenMP or both. The benefit may be briefly summarized as follows: MPI has distribution across orbitals, hence a basic advice is that you get performance scaling with # of cores == # of atoms (rule of thumb) OpenMP also distributes across orbitals,